One API key. Reliable access to the models you need.
ModelFerry gives you rate-limited API keys for Minimax-M3 and Kimi K3 Coding, with more models to come — provisioned in minutes, no infrastructure to manage.
curl https://api.modelferry.com/v1/chat/completions \
-H "Authorization: Bearer $MODELFERRY_API_KEY" \
-H "Content-Type: application/json" \
-d '{ "model": "minimax-m3", "messages": [...] }'How it works
Create a free account
Sign up with your email — no credit card required to get started.
Get your API key instantly
Verify your email and a live key on the Free plan is provisioned for you automatically.
Call the gateway
Drop your key into any HTTP client and start making requests right away.
Simple, usage-based pricing
Start free. Upgrade any time from your dashboard — your existing API key keeps working through an upgrade, just with higher limits.
Free
Get an API key in minutes, no card required.
5,000 requests / 1h
- 500 requests/min
- 5,000 requests/hour
- Access to Minimax-M3
- Community support
Designer
For individual projects with steady traffic.
10,000 requests / 1h
- 1,000 requests/min
- 10,000 requests/hour
- Access to Minimax-M3
- Email support
Builder
For production apps with growing usage.
50,000 requests / 1h
- 5,000 requests/min
- 50,000 requests/hour
- Access to Minimax-M3
- Priority email support
Architect
For high-volume products that can't afford limits.
500,000 requests / 1h
- 50,000 requests/min
- 500,000 requests/hour
- Access to Minimax-M3
- Priority support
Ronin
Get started with Kimi K3 Coding.
5,000 requests / 1h
- 500 requests/min
- 5,000 requests/hour
- Access to Kimi K3 Coding
- Community support
Shinobi
For individual coding projects with steady traffic.
10,000 requests / 1h
- 1,000 requests/min
- 10,000 requests/hour
- Access to Kimi K3 Coding
- Email support
Samurai
For production coding workloads with growing usage.
50,000 requests / 1h
- 5,000 requests/min
- 50,000 requests/hour
- Access to Kimi K3 Coding
- Priority email support
Shogun
For high-volume coding products that can't afford limits.
500,000 requests / 1h
- 50,000 requests/min
- 500,000 requests/hour
- Access to Kimi K3 Coding
- Priority support
Starter
Get started with the full Kimi model family.
5,000 requests / 1h
- 500 requests/min
- 5,000 requests/hour
- Access to kimi-k2.5, kimi-k2.6, kimi-k2.7-code, kimi-k3
- Community support
Growth
For individual projects with steady traffic across the Kimi family.
10,000 requests / 1h
- 1,000 requests/min
- 10,000 requests/hour
- Access to kimi-k2.5, kimi-k2.6, kimi-k2.7-code, kimi-k3
- Email support
Scale
For production apps with growing usage across the Kimi family.
50,000 requests / 1h
- 5,000 requests/min
- 50,000 requests/hour
- Access to kimi-k2.5, kimi-k2.6, kimi-k2.7-code, kimi-k3
- Priority email support
Enterprise
For high-volume products that can't afford limits on the Kimi family.
500,000 requests / 1h
- 50,000 requests/min
- 500,000 requests/hour
- Access to kimi-k2.5, kimi-k2.6, kimi-k2.7-code, kimi-k3
- Priority support
Starter
Get started with the full GLM model family.
5,000 requests / 1h
- 500 requests/min
- 5,000 requests/hour
- Access to glm-5.1, glm-5.2
- Community support
Growth
For individual projects with steady traffic across the GLM family.
10,000 requests / 1h
- 1,000 requests/min
- 10,000 requests/hour
- Access to glm-5.1, glm-5.2
- Email support
Scale
For production apps with growing usage across the GLM family.
50,000 requests / 1h
- 5,000 requests/min
- 50,000 requests/hour
- Access to glm-5.1, glm-5.2
- Priority email support
Enterprise
For high-volume products that can't afford limits on the GLM family.
500,000 requests / 1h
- 50,000 requests/min
- 500,000 requests/hour
- Access to glm-5.1, glm-5.2
- Priority support
Starter
Get started with the full DeepSeek model family.
5,000 requests / 1h
- 500 requests/min
- 5,000 requests/hour
- Access to deepseek-v4-flash, deepseek-v4-pro
- Community support
Growth
For individual projects with steady traffic across the DeepSeek family.
10,000 requests / 1h
- 1,000 requests/min
- 10,000 requests/hour
- Access to deepseek-v4-flash, deepseek-v4-pro
- Email support
Scale
For production apps with growing usage across the DeepSeek family.
50,000 requests / 1h
- 5,000 requests/min
- 50,000 requests/hour
- Access to deepseek-v4-flash, deepseek-v4-pro
- Priority email support
Enterprise
For high-volume products that can't afford limits on the DeepSeek family.
500,000 requests / 1h
- 50,000 requests/min
- 500,000 requests/hour
- Access to deepseek-v4-flash, deepseek-v4-pro
- Priority support
Starter
Get started with the General model family.
5,000 requests / 1h
- 500 requests/min
- 5,000 requests/hour
- Access to qwen3.5-397b, gemma4-31b, gpt-oss-120b
- Community support
Growth
For individual projects with steady traffic across the General family.
10,000 requests / 1h
- 1,000 requests/min
- 10,000 requests/hour
- Access to qwen3.5-397b, gemma4-31b, gpt-oss-120b
- Email support
Scale
For production apps with growing usage across the General family.
50,000 requests / 1h
- 5,000 requests/min
- 50,000 requests/hour
- Access to qwen3.5-397b, gemma4-31b, gpt-oss-120b
- Priority email support
Enterprise
For high-volume products that can't afford limits on the General family.
500,000 requests / 1h
- 50,000 requests/min
- 500,000 requests/hour
- Access to qwen3.5-397b, gemma4-31b, gpt-oss-120b
- Priority support